Citations

Abstract INTRODUCTION Recurrent somatic mutations in the metabolic enzymes isocitrate dehydrogenase 1 and 2 (IDH1/2) confer gain-of-function activity in cancer cells, resulting in accumulation of the oncometabolite, D-2-hydroxyglutarate (2-HG). High levels of 2-HG result in epigenetic changes and impaired cellular differentiation. IDH mutations have been identified in multiple solid tumors and hematologic malignancies. Approximately 6-10% and 9-13% of adults with acute myeloid leukemia (AML) carry mutations in IDH1 (mIDH1) or IDH2 (mIDH2), respectively. AG-120 is a first-in-class, oral, potent, reversible, selective inhibitor of the IDH1 mutant enzyme under evaluation in multiple ongoing single agent and combination clinical trials [NCT02074839, NCT02073994, NCT02632708, NCT02677922]. This is the first report of IDH1 mutation clearance assessed by variant allele frequency (VAF) analysis using next-generation sequencing (NGS) in patients treated on the dose escalation portion of the first-in-human phase 1 study [NCT02074839]. METHODS Patients with advanced mIDH1-positive hematologic malignancies received AG-120 as a single agent orally once daily (QD) or twice daily (BID) continuously in 28-day cycles. Primary objectives were determination of safety, maximum tolerated dose (MTD), and selection of a dose schedule for expansion cohorts and future studies. Secondary objectives included clinical activity assessed by investigators using modified 2003 International Working Group Criteria in AML. Determination of mIDH1 VAF was performed using the FoundationOne® Heme test on mononuclear cells from the bone marrow or peripheral blood at screening and subsequent time points on study. This NGS assay reports IDH1 mutations for samples with VAF ≥1%, with median coverage 500X. Patients with IDH1 mutational clearance (IDH1-MC) were defined as having mIDH1 at baseline and at least one on-study sample with no reported mIDH1. RESULTS Seventy-eight patients were treated in the dose escalation portion, which is now closed to enrollment. As of the data cut-off of May 12, 2016, the median duration on treatment was 3.2 months and 9 (11.5%) patients remain on treatment, with an additional 8 (10.3%) patients transitioned to stem cell transplant. Doses ranged from 300-1200 mg QD with 1 cohort at 100 mg BID. Though the MTD was not reached, the recommended phase 2 dose was determined to be 500 mg QD. The majority of adverse events (AEs) were grade 1 and 2, the most common (≥30%) being diarrhea, fatigue, and nausea; the most common grade ≥3 AEs (≥15%) were febrile neutropenia, anemia, leukocytosis and pneumonia. The most common serious AEs were febrile neutropenia (16.7%) and pneumonia (12.8%). The overall response rate (ORR) was 38.5% (n=30), with 17.9% (n=14) achieving a complete remission (CR). Longitudinal mIDH1 VAF data were available for 51 patients; of these, 22% (n=11) achieved a CR. IDH1-MC was observed in 27.3% (3/11) patients who achieved CR (Figure 1). In contrast, only 1/40 patients who did not achieve CR experienced IDH1-MC. This occurred in a patient with an initially low mIDH1 VAF who had clinical progression despite IDH1-MC (Figure 1, bottom right). In all 3 patients with CR who achieved IDH1-MC, an initial increase in mIDH1 VAF, or early peak, was observed prior to IDH1-MC, suggesting that early clonal expansion might have occurred as part of the mechanism of action of AG-120. CONCLUSION This is the first demonstration that treatment with single agent AG-120 can result in mIDH1 clearance as determined by NGS. Further analysis of the mutational profiles is planned. AG-120, a potent, selective, oral inhibitor of mIDH1 continues to demonstrate a well-tolerated safety profile in patients with advanced hematologic malignancies, and induced objective single-agent durable responses. The data continue to support the efficacy and safety of single agent AG-120 and provide evidence that the underlying biology of the disease is altered by treatment. Figure 1. VAF analysis using FoundationOne® Heme NGS assay in 4 AML patients with IDH1-MC treated with AG-120 Figure subscript: Y-axis is mIDH1 VAF, x-axis is days on treatment. Text indicates investigator-assessed clinical response. CR, complete remission; CRi, CR with incomplete neutrophil recovery; CRp, CR with incomplete platelet recovery; R/R, relapsed/refractory; SD, stable disease; PD, progressive disease; *post-transplant sample Figure 1. VAF analysis using FoundationOne® Heme NGS assay in 4 AML patients with IDH1-MC treated with AG-120. / Figure subscript: Y-axis is mIDH1 VAF, x-axis is days on treatment. Text indicates investigator-assessed clinical response. CR, complete remission; CRi, CR with incomplete neutrophil recovery; CRp, CR with incomplete platelet recovery; R/R, relapsed/refractory; SD, stable disease; PD, progressive disease; *post-transplant sample Disclosures DiNardo: Daiichi Sankyo: Other: advisory board, Research Funding; Novartis: Other: advisory board, Research Funding; Abbvie: Research Funding; Celgene: Research Funding; Agios: Other: advisory board, Research Funding. de Botton:Novartis: Consultancy; Agios: Consultancy, Research Funding; Celgene: Consultancy, Honoraria, Membership on an entity's Board of Directors or advisory committees; Pfizer: Consultancy; Servier: Consultancy; Pierre Fabre: Consultancy. Stein:Agios: Other: advisory board; Celgene: Other: advisory board; Novartis: Other: advisory board. Roboz:Cellectis: Research Funding; Agios, Amgen, Amphivena, Astex, AstraZeneca, Boehringer Ingelheim, Celator, Celgene, Genoptix, Janssen, Juno, MEI Pharma, MedImmune, Novartis, Onconova, Pfizer, Roche/Genentech, Sunesis, Teva: Consultancy. Pollyea:Alexion: Other: advisory board; Pfizer: Other: advisory board, Research Funding; Ariad: Other: advisory board; Glycomimetics: Other: DSMB member; Celgene: Other: advisory board, Research Funding. Fathi:Bexalata: Other: Advisory Board participation; Agios Pharmaceuticals: Other: Advisory Board participation; Seattle Genetics: Consultancy, Other: Advisory Board participation, Research Funding; Merck: Other: Advisory Board participation; Celgene: Consultancy, Research Funding. Altman:Syros: Honoraria; BMS: Honoraria; Janssen Pharmaceuticals: Honoraria; Novartis: Honoraria. Flinn:Janssen: Research Funding; Pharmacyclics LLC, an AbbVie Company: Research Funding; Gilead Sciences: Research Funding; ARIAD: Research Funding; RainTree Oncology Services: Equity Ownership. Foran:novartis: Honoraria; Millennium Pharmaceuticals, Inc.: Research Funding; karyopharm: Honoraria; medscape: Honoraria; pfizer: Honoraria; boehringer ingelheim: Research Funding; agios: Research Funding; Cellerant: Research Funding. Pigneux:Sunesis: Consultancy, Honoraria; Agios: Consultancy, Honoraria. Kantarjian:ARIAD: Research Funding; Bristol-Myers Squibb: Research Funding; Amgen: Research Funding; Pfizer Inc: Research Funding; Delta-Fly Pharma: Research Funding; Novartis: Research Funding. Liu:Agios Pharmaceuticals, Inc.: Employment, Equity Ownership. Attar:Agios Pharmaceuticals, Inc.: Employment, Equity Ownership. Sacolick:Agios Pharmaceuticals, Inc.: Employment, Equity Ownership. Yen:Agios Pharmaceuticals, Inc.: Employment, Equity Ownership. Hurov:Agios Pharmaceuticals, Inc.: Employment, Equity Ownership. Choe:Agios Pharmaceuticals, Inc.: Employment, Equity Ownership. Wu:Agios Pharmaceuticals, Inc.: Employment, Equity Ownership. Stone:Amgen: Consultancy; Sunesis Pharmaceuticals: Consultancy; Pfizer: Consultancy; Roche: Consultancy; Abbvie: Consultancy, Membership on an entity's Board of Directors or advisory committees; Agios: Consultancy; Celator: Consultancy; Celgene: Consultancy, Membership on an entity's Board of Directors or advisory committees; Karyopharm: Consultancy; Novartis: Consultancy; Jansen: Consultancy; ONO: Consultancy; Juno Therapeutics: Consultancy; Merck: Consultancy; Seattle Genetics: Consultancy; Xenetic Biosciences: Consultancy.

Abstract Introduction eIF5A is the only known protein to be modified by hypusination and is highly conserved across species. Hypusinated eIF5A, the predominant form in normal and cancer cells, is involved in cell survival and inflammatory pathway activation. siRNAs targeting eIF5A inhibit NF-kB activation and reduce pro-inflammatory cytokine production. Accumulation of the unhypusinated lysine form of eIF5A is associated with apoptosis. Mutants of eIF5A that cannot be hypusinated (e.g. eIF5AK50R) are pro-apoptotic in vitro and have anti-tumoral activity in vivo in multiple cancer types including melanoma and lung cancer. SNS01-T is a novel therapeutic with a dual mechanism of eIF5A modulation: inducing cell death via siRNA-mediated inhibition of hypusinated eIF5A while simultaneously causing over-expression of pro-apoptotic eIF5AK50R via a DNA plasmid with a B-cell promoter to induce tumor cell death. SNS01-T significantly inhibited tumor growth and increased survival in mouse models of myeloma (MM), mantle cell and diffuse large B-cell lymphoma. The phase 1-2 study of SNS01-T has completed 4 planned dosing cohorts 0.0125, 0.05, 0.2 and 0.375 mg/kg twice weekly IV for 6 weeks in pts with refractory B-cell cancers. Methods PK and PD secondary endpoints included characterization of PK by measuring pExp5A plasmid DNA and eIF5A siRNA in blood and bone marrow (BM), assessing potential immunogenicity of SNS01-T by measuring serum concentrations of antibodies against SNS01-T nanoparticles, and measuring serum concentrations of select proinflammatory cytokines by enzyme-linked immunosorbent assay in serum and plasma samples. Blood PK timepoints were 30 minutes before the first infusion and at 30 minutes, 2, 6, and 24 hours after the first infusions on Week 1, Week 3, and Week 6 and at the 4, 8, and 12 week visits after the last infusion; BM samples were collected 1 day after the final infusion. Serum and plasma PD sampling timepoints were 30 minutes before and at 2, 6, and 24 hours after the first infusion on Week 1, Week 3, and Week 6, and at 4 weeks after the final infusion. Cytokines assayed included TNF-α, IFN-α, IFN-ß, IFN-g, CXCL1, IL-1ß, IL-2, IL-4, IL-5, IL-6, IL-10, and IL-12. Results Table 1 shows data available for interpretation as of August 2014. The remaining samples are under analysis and will be presented. Abstract 2121. Table 1:Data available for interpretation (number of patients)SNS01-T DoseOverall(n=18)0.0125 mg/kg(n=6)0.05 mg/kg(n=4)0.2 mg/kg(n=4)0.375 mg/kg(n=4)Blood PK DNA and RNA156441Bone marrow PK82321Serum antibodies to SNS01-T nanoparticles186444Serum and plasma cytokines156441 Plasmid and siRNA blood levels generally peaked 30 minutes post-dosing at weeks 1, 3 and 6 of dosing. Both plasmid and siRNA exhibited rapid clearance from the blood, with levels dropping to near pre-dosing levels within 24 hours of administration. pExp5A plasmid DNA was detectable in the bone marrow of 2 pts at cohort 1, 2 at cohort 2, 1 at cohort 3 and 1 at cohort 4. eIF5A siRNA was not detectable in bone marrow. No antibodies to SNS01-T nanoparticles were detected at any timepoint at any dose level. Cytokines remained within the expected range of inter-patient variability, similar to baseline across all timepoints at the first 2 dose levels. At dose level 3, levels of IL-6, IL-8 and TNF-α in particular increased at the 2 and 6 hour timepoints but had recovered to baseline levels 24-hours post dosing. This effect was more pronounced at the first infusion. Conclusions PCR analysis demonstrated the presence of both plasmid DNA and siRNA components of SNS01-T in blood at all dose levels, with a dose-dependent increase in plasmid copy number. Plasmid DNA was also detected in bone marrow collected 24 hours after the final infusion of SNS01-T. Pro-inflammatory cytokines did increase within hours of infusion but returned to baseline within 24 hours, synchronous with the clinical infusion reactions (see Abstract 70148). No evidence of an anti-SNS01-T antibody response was observed in any subject. Phase 2 trials are planned. Disclosures Bexon: Senesco: Consultancy. Craig:Senesco: PI Other. Siegel:Senesco: PI Other. Bensinger:Senesco: PI Other. Novitzky:Senesco: PI Other. McDonald:Senesco: PI Other. Gutierrez:Senesco: PI Other. Libby:Senesco: PI Other. van Rhee:Senesco: PI Other. Heidel:Senesco: Consultancy. Thompson:Senesco: Consultancy, Employment, Equity Ownership, Membership on an entity's Board of Directors or advisory committees, Patents & Royalties, Research Funding. Barranco:Senesco: Consultancy. Taylor:Senesco: Research Funding. Browne:Senesco: Employment. Kurman:Senesco: Consultancy. Lust:Senesco: PI Other. Dondero:Senesco: Employment.

Abstract Abstract 684 Background. Obesity has been associated with an increased risk of developing of lymphoma, particularly DLBCL. Genetic variation in several energy balance genes, including leptin (LEP), its receptor LEPR, ghrelin (GHRL) and neuropeptide Y (NPY) have been shown to modulate risk of lymphoma, although their impact on survival is not known. In contrast, recent studies suggest that patients either overweight (Body Mass Index (BMI), 25 to<30 kg/m2) or obese (≥30 kg/m2) may have a superior prognosis in DLBCL, although its impact in FL and in the current treatment era is not known. We evaluated the association of BMI at the time of diagnosis, 2 years prior to diagnosis and at ages 18, 35 and 50 years, along with germline genetic variation in selected energy balance genes with event-free (EFS) and overall (OS) survival in DLBCL and FL. Methods. We evaluated these associations in a prospective cohort of newly diagnosed DLBCL (N=764) and FL (N=649) patients enrolled at the Mayo Clinic and University of Iowa from 2002–2009 as part of the Molecular Epidemiology Resource. Peripheral blood DNA, clinical data and self-reported historical weights were collected at enrollment. Patients received standard disease treatment strategies, and were systematically followed for disease progression, re-treatment, and death. Single nucleotide polymorphisms (SNPs) were genotyped as part of a custom Illumina iSelect panel. The most prevalent homozygous genotype was used as the reference group, and each SNP was modeled as having a log-additive effect in the regression model. Cox proportional hazards regression was used to estimate hazard ratios (HRs) and 95% confidence intervals (CI) for the association of the obesity measures and SNPs with outcome, adjusted for treatment class and subtype specific prognostic factors (IPI, FLIPI). Results. For DLBCL patients (median age 63 years, range 18–92) the median follow-up was 47 months (range 0–101); 58% had an IPI 0–1 risk disease; and 87% received immunochemotherapy. For FL patients (median age 60 years, range 25–94), the median follow-up was 60 months (range 11–110); 24% had FLIPI score 3–5; and main initial treatments were observation (32%), rituximab (R)-monotherapy (12%), CVP +/−R (19%), and anthracycline +/−R (22%). The main results for BMI and prognosis are reported in the Table. For DLBCL, obese patients at diagnosis had better EFS and OS; this association was weaker for BMI at age 50 and 35 years. In contrast, for FL, overweight and obese patients at diagnosis had a poorer EFS and OS, and this was also observed for BMIs two years before diagnosis and at ages 50 and 35 years. In FL, the LEP SNP rs2167270 was associated with inferior EFS (HR=1.25; 95%CI 1.05–1.49) and OS (HR=1.28; 95%CI 0.90–1.83). The NPYSNP rs16147 was associated with inferior FL OS (HR=1.50; 95%CI 1.03–2.17) but was not associated with EFS. None of the energy balance and metabolism SNPs were associated with DLBCL EFS or OS. Conclusions. Higher BMI prior to and at diagnosis, as well as SNPs in LEP and NPY,were significantly associated with FL prognosis, with inferior outcomes in overweight and obese patients. These novel findings in the modern treatment era suggest an important role of obesity pathways in FL prognosis. In DLBCL, we confirm the recent findings of improved survival in DLBCL in obese patients, and extend this to the modern treatment era and BMI earlier in life. Disclosures: No relevant conflicts of interest to declare.

INTRODUCTION: Bariatric surgery (BRS) is an effective weight loss method that should, in theory, improve many gastrointestinal, liver, and pancreatic (GILP) diseases seen with obesity. At the same time, there is an increased morbidity related to BRS’s gastrointestinal alterations. Thus, we examined the short-term impact of BRS type on admission rates of various GILP diseases. METHODS: We queried the 2012-2014 Nationwide Readmission Database to identify obese patients who underwent vertical sleeve gastrectomy (VSG), Roux-en-Y gastric bypass (RYGB), or, as a control, hernia repair surgery (HR). Our primary aim was to compare rates of GILP diseases within 6 months post- and 6 months pre-surgery in the VSG, RYGB, and HR cohorts. In a time to event analysis we evaluated readmission rates up to 11 months post-BRS. RESULTS: We identified a total of 140,103 patients in the VSG cohort, 132,253 patients in the RYGB cohort, and 12,436 patients in the HR cohort. HR patients were slightly older with a higher percentage of Medicare insurance carriers in contrast to VSG and RYGB (Table 1). RYGB was associated with a 33-fold increased risk of intestinal obstructions compared to pre-RYGB ( P < 0.001). Other notably increased GILP diseases after RYGB from highest to lowest risk (Table 2): C. difficile infection [adjusted odds ratio (aOR): 12.52, P < 0.001], GI hemorrhage (aOR: 10.95, P < 0.001), biliary disease (aOR: 4.15, P < 0.001) and functional/motility disorders (aOR: 3.57, P < 0.001). In contrast, post- versus pre-VSG patients were most frequently admitted for acute pancreatitis (aOR: 6.26, P < 0.001), followed by C. difficile infection (aOR: 4.86, P < 0.001), biliary disease (aOR: 3.14, P < 0.001), noninfectious gastroenteritis (aOR: 3.05, P < 0.001) and functional/motility disorders (aOR: 2.74, P < 0.001). No increase in admissions was observed in the control HR cohort for any of these diseases. In our time to event analysis, the probability of being admitted was most notable for intestinal obstructions within 30 days after RYGB while functional disorders predominate afterwards (Figure 1). Similarly, patients were more frequently admitted for functional disorders following VSG. CONCLUSION: The type of BRS has a differential impact on GILP admissions after surgery. There may be a limited impact of weight loss on functional disorders that seem to dominate irrespective of BRS type. This data can help better inform patients on the choice of surgery to pursue, however more data are warranted.

INTRODUCTION: There are multiple assays for infliximab (IFX) trough level (IFX-TL) and antibody (ATI) measurement. The aims of this study are to examine the correlation and outcomes of IFX-TL and ATI measured with liquid chromatography-tandem mass spectrometry (LC-MS/MS) and electrochemiluminescence (ECLIA) in inflammatory bowel disease (IBD) patients. METHODS: This was a prospective study of IFX-treated IBD patients undergoing drug monitoring with an ECLIA panel (Esoterix Laboratories). Residual serum samples were used to simultaneously quantify IFX-TL via LC-MS/MS and ATI via an in-house ECLIA (iECLIA) (Mayo Clinic). Both LC-MS/MS and iECLIA have been previously validated. Primary outcomes: IFX-TL and ATI comparison between the two assays. Secondary outcomes: association of IFX-TL and ATI with clinical remission, mucosal healing (MH) (absence of ulcers in Crohn's disease and Mayo endoscopic score ≤1 for ulcerative colitis) and normal serum C-reactive protein (CRP ≤8 mg/L). RESULTS: A total of 151 patients were included (median age 32 years (12-84), 45.7% female) (Table 1). Median IFX-TL was 7 mcg/mL (IQR: 1.3, 19.4) and 6 mcg/mL (IQR: 0.9, 20) via LC-MS/MS and ECLIA, respectively (Spearman correlation coefficient r = 0.975). Median IFX-TL via LC-MS/MS was not significantly different between patients in clinical remission, response and no response (5.7 vs. 8.4 vs. 2.3 mcg/mL, respectively; P = 0.46), but significantly higher in patients who had a normal CRP (11.1 vs. 1.4 mcg/mL; P < 0.001) and in those who achieved MH (13.1 mcg/mL vs. 3.9 mcg/mL; P = 0.01). ATI was detected in 13/142 (9.2%) via iECLIA and 39/151 (25.8%) via ECLIA. The frequency of ATI detection via iECLIA was lower in patients in clinical remission or response than no response (7.3% vs. 9.3% vs. 15.4%; P = 0.68). On the contrary, the frequency of ATI detection via ECLIA was higher in patients with clinical remission (35.7%) than response (18.7%) and comparable to patients who had no clinical response (40%) ( P = 0.05) (Table 2). The frequency of ATI was not significantly different between patients who did and did not achieve MH or had normal CRP using both assays (Table 2). CONCLUSION: IFX-TL was comparable between LC-MS/MS and ECLIA assays. IFX-TL correlated with MH and normal CRP but not clinical remission. Frequency of ATI detection was not significantly associated with outcomes via both methods. ATI detected via iECLIA showed a trend towards correlation with clinical outcomes compared to ECLIA.

INTRODUCTION: Combination immunomodulator (IMM) with infliximab (IFX) has been shown to be superior to IFX monotherapy therapy in the management of inflammatory bowel disease (IBD). It is unclear if this is related to IMM solely or higher IFX trough level (IFX-TL). The aims of this study are to examine the correlation of IFX-TL with clinical outcomes of IBD patients and to evaluate whether IFX-TL or combination IMM therapy correlate best with mucosal healing (MH). METHODS: This is a retrospective study of consecutive IFX treated IBD patients undergoing drug monitoring with liquid chromatography-tandem mass spectrometry (LC-MS/MS). IMM included either thiopurines or methotrexate. Primary analyses included: comparing IFX-TL with clinical remission, MH (absence of mucosal ulcers in Crohn's disease (CD) and Mayo endoscopic score ≤1 for ulcerative colitis (UC)) and normal c-reactive protein (CRP ≤8 mg/L). Secondary analyses included determining an IFX-TL cut off to achieve MH and evaluating the predictors of MH (IMM vs. IFX-TL). RESULTS: A total of 151 patients were included (67% CD, 24% UC, 6% indeterminate colitis, 3% pouchitis) (Table 1). The median IFX-TL was 7 mcg/mL (IQR: 1.3, 19.4). The distribution of IFX-TL was not significantly different between patients in clinical remission, response and no response (median: 5.7 vs. 8.4 vs. 2.3 mcg/mL, respectively; P = 0.46). The median IFX-TL was significantly higher in patients who had a normal CRP (11.1 vs. 1.4 mcg/mL; P < 0.001) and in those who achieved MH (13.1 mcg/mL vs. 3.9 mcg/mL; P = 0.01). An IFX-TL ≥7.5 mcg/mL predicted MH with a sensitivity of 67% and specificity of 64% (AUC = 0.67; Figure 1). Combination therapy was associated with higher IFX-TL (median: 15.8 vs. 4.6 mcg/mL; P < 0.001). In a multivariable model to predict MH based on IFX-TL (continuously), combination therapy (yes vs no), dosing frequency (≥8 vs <8 weeks), and albumin (<3.5 vs ≥3.5), higher IFX-TL was associated with higher likelihood of MH (OR for 5-mcg/mL increase: 1.18, 95% CI 1.01-1.39; P = 0.04). In this model, IMM combination was not significant (OR: 0.79, P = 0.66), neither was dosing frequency (OR: 1.44, P = 0.48). Low albumin showed lower likelihood of MH, but this was not statistically significant (OR 0.14, 95% CI: 0.02-1.16, P = 0.07). CONCLUSION: IFX-TL was associated with normal CRP and MH but not clinical remission in this cohort of IBD patients. IFX-TL and not combination IMM therapy was independently associated with MH.

INTRODUCTION: The use of proton-pump inhibitors (PPIs) has been associated with various adverse effects including increased risk of major adverse cardiovascular events, community acquired clostridium difficile infection, chronic kidney disease and dementia. Some studies over the past decade have suggested an association between PPIs use and ischemic stroke. METHODS: We pooled the data after comprehensive literature search from the PubMed, Embase, Cochrane Library and Scopus databases using ‘proton pump inhibitors’ AND ‘stroke’ as the keywords. RevMan was used for the entry and analysis of data. A random effects model was created given the significant heterogeneity of the included studies and a funnel plot was made to assess the publication bias. Our primary objective was to study whether all-comers patients on PPIs are at an increased risk of stroke. Sub-group analysis was also done to investigate the association between patients on dual anti-platelet therapy [DAPT] (aspirin and clopidogrel) with or without PPIs and stroke. RESULTS: After searching 774 titles, 25 titles were selected to be included in the systematic review and 15 articles for the meta-analysis. Our pooled analysis of 155,406 patients (46,652 patients on PPI and 101,892 patients without PPI found a positive association between the use of PPI and stroke (odds ratio = 1.28, 95% confidence interval = 1.12, 1.48, P = 0.001) [Figure 1]. Given the significant heterogeneity of the studies (I2= 60%) a sub-group analysis was done which also suggested an increased risk of stroke associated with PPI use among patients on dual antiplatelet therapy (OR: 1.43, 95% CI: 1.26, 1.66, p< 0.001, I2= 11%) [Figure 2]. Analysis of the included randomized controlled trials revealed an OR: 1.47, 95% CI: 0.66, 3.25 but the results were not statistically significant (p< 0.81) [Figure 3]. CONCLUSION: In a pooled analysis of over 155,000 patients we found a positive association between the use of PPIs and stroke. Our results are limited by significant heterogeneity and retrospective study design of the observational studies. Rigorous clinical trials are needed to establish the safety of PPIs in relation to stroke, especially among patients on DAPT. Until definitive data emerges, patients on DAPT are probably best served by being off PPIs or being on alternative acid suppression regimen, unless there is a compelling indication for PPI use.